ZrO2 and ZrO2SiC particle reinforced MoSi2 matrix composites

Materials Science and Engineering, A 155 (1992) 259-266

Z r O 2 and Z r O 2 - S i C

259

particle reinforced M o S i 2 matrix composites

J. J. Petrovic, A. K. Bhattacharya, R. E. Honnell and T. E. Mitchell Ceramic Science and Technology Group, Los Alamos National Laboratory, Los Alamos, NM 87545 (USA)

R. K. Wade Arizona Materials" Laboratory, University of Arizona, Tucson, A Z 85721 (USA)

K. J. McClellan Department of Materials Science and Engineering, Case Western Reserve University, Cleveland, OH 44106 (USA)

Abstract ZrO~-MoSi~ and (ZrO2-SiC)-MoSi2 composites were fabricated by hot pressing and hot pressing-hot isostatic pressing at 1700 °C. No reactions between ZrO2, SiC and MoSi 2 were observed. An amorphous silica glassy phase was present in all composites. Composites with unstabilized ZrO 2 particles exhibited the highest room temperature fracture toughness, reaching a level three times that of pure MoSi 2. Both the room temperature toughness and 1200 °C strength of (ZrO2-SiC)-MoSi2 composites were higher than those of ZrO2-MoSi 2 composites, indicating beneficial effects of combined reinforcement phases. Low strength levels were observed at 1400 °C as a result of the presence of the silica glassy phase. Elimination of glassy phases and refinements in microstructural homogeneity are processing routes important to the optimization of the mechanical properties of these types of composites.

1. Introduction

Transformation toughening has been shown to be a very important toughening mechanism in zirconiabased materials [1-6]. In this mechanism, metastable tetragonal zirconia particles transform to monoclinic zirconia via a martensitic transformation, either spontaneously on cooling or in the vicinity of crack tip stress fields. The volume change associated with the transformation produces the observed toughening effects. The transformation is sensitive to alloying additions to the ZrO2 (the most important alloying species are MgO, CaO, CeO2 and Y203), and particle size of the tetragonal phase. Zirconia has also been added to alumina in order to increase the fracture toughness of ZrO2-AI203 composites [7-9[. These composites are referred to as zirconia-toughened alumina (ZTA). Some investigators have also added SiC whiskers to zirconia-toughened materials in order to enhance high temperature mechanical properties [ 10, 11 ]. Recent work has demonstrated that zirconia transformation toughening may also be employed to toughen MoSiz-based materials significantly [12, 13], when ZrO 2 particles are present as a dispersed phase in an MoSi 2 matrix. The chemical species ZrO2 and MoSi2 are stable with each other under inert condi0921-5093/92/$5.00

tions, and the presence of the ZrO2 does not significantly degrade composite oxidation resistance. In the present investigation, we compare the fabrication, microstructures, and mechanical properties of Z r O 2 particle-MoSi 2 matrix and combined ZrO,-SiC particle-MoSi 2 matrix composites.

2. Experimental details

2.1. Fabrication 2.1.1. ZrO 2-MoSi 2 composites A range of commercial ZrO2 powders from the Tosoh Corporation was employed. Powders containing 0, 2, 2.5, 3, 4 and 8 mol.% Y203 w e r e used. The 0 tool.% Y203 material is unstabilized pure ZrO2. The range of 2-4 mol.% Y203 constitutes partially stabilized ZrO2. The 8 tool.% Y203 is fully stabilized ZrO2. All Tosoh powders were high purity, with an average particle size of 0.3/~m. MoSi 2 powder used was commercial powder from the Alfa Corporation. This powder was screened to - 4 0 0 mesh prior to mixing with the various ZrO 2 powders. ZrO2 and MoSi2 powders were co-dispersed in an aqueous media at a pH of 2.5. Mechanical stirring and ultrasonification were employed. The solids © 1992 - Elsevier Sequoia. All rights reserved,

260

J.J. Petrovic et al. / ZrO_,-MoSi2and (ZrOySiC)-MoSi 2 composites

loading was 60 wt.%. The powder co-dispersion was slip cast into a plaster of paris mold, and the slip cast body crushed into - 1 0 0 mesh feed powder for hot pressing. Hot pressing consolidation of composites was performed at 1700°C and 32 MPa pressure, using grafoil-lined graphite dies and an argon atmosphere. Hot pressed composites were 94%-95% dense, and contained 20 vol.% ZrO2 particles.

ceramic bend fixture, with the specimens heated with an MoSi 2 element mechanical testing furnace. All mechanical tests were performed on an Instron test frame.

3. Results 3.1. Microstructures

2.1.2. (ZrO 2-SIC) - M o S i 2 composites

Tosoh zirconia powders containing 0, 2.5 and 8 mol.% Y203 were employed. Alpha SiC (0.5/zm particle size) SiC powder from the H. C. Starck Corporation was used. The MoSi 2 powder was H. C. Starck grade C, with a particle size of 0.9-1.3 /zm, and an indicated oxygen content of 2 wt.%. ZrO2, SiC and MoSi 2 powders were roll blended dry with WC balls. The blend was then dry milled using a Megapact vibratory mill with WC balls. The powder was contained in a polyethylene jar, and milling time was 1 h. Milled powders were hot pressed at 1700 °C and 3 2 M P a pressure for 30min, using grafoil-lined graphite dies and an argon atmosphere. Hot pressed composites were then treated by hot isostatic pressing (HIP) in a containerless manner at 1700°C and 207 MPa for 30 min in argon. HIP composites were 100% dense, and contained 10 vol.% ZrO 2 and 10 vol.% SiC particles.

Representative SEM micrographs of the polished microstructures of 20 vol.% ZrO2-MoSi2 and (10vol.%ZrO2-10vol.%SiC)-MoSi 2 composites are shown in Fig. 1. As may be seen, the (ZrOz-SiC)-MoSi2 composites exhibited significantly more homogeneous microstructures than the ZrO2-MoSi 2 composites. This difference is due to the difference in fabrication procedures for the two types of composites. It suggests that high energy vibratory milling followed by hot

2.2. Microstructures

Composite microstructures were examined with light optical X-ray diffraction (XRD), scanning electron microscopy (SEM) and transmission electron microscopy (TEM) techniques. Light optical and SEM techniques were employed to assess microstructural homogeneity. XRD was performed to determine phases present after fabrication. TEM was used to describe composite substructures. T E M studies of substructures were performed using a Philips CM30 transmission electron microscope at 300 kV. 2.3. Mechanical properties

Microhardness indentation techniques were employed to determine the room temperature fracture toughness of the composites. Vickers' indentations were used and the approach of Anstis et al. [14] was adopted to calculate fracture toughness values. Four-point bend strength was measured as a function of temperature. Bend specimen dimensions were 2.54 mm x 5.08 mm x 25.4 mm. Four-point bend test fixtures had an outer span of 19 mm and an inner span of 9.5 mm. A hardened tool steel self-aligning bend fixture was employed at room temperature. Elevated temperature bend tests were conducted using a

Fig. 1. Comparison of the polished microstructures of (a) 20vol.%ZrO2-MoSiz and (b) (10vol.%ZrO2-10vol.%SiC)MoSi2composites.

J. J. Petrovic et al.

/

ZrO,-MoSi, and (ZrO:-SiC)-MoSi: composites

261

pressing-HIP consolidation is a preferred fabrication methodology to aqueous co-dispersion of powders with only hot pressing consolidation. For the case of the ZrO2-MoSi 2 composites, the ZrO 2 particles tended to be rather inhomogeneously clustered between regions of MoSi~ phase, rather than discretely dispersed as individual particles. For the (ZrO2-SiC)M o S i 2 composites, the extent of clustering was reduced, though not totally eliminated.

3.2. X-ray diffraction phase analysis XRD studies of the ZrO2-MoSi 2 composites after fabrication consolidation at 1700°C indicated the presence of only Z r O 2 and M o S i 2 phases. The crystal structure of the Z r O 2 phase was dependent on the amount of Y20~ addition. The Z r O 2 with 0 tool.% Y203 was monoclinic, and the Z r O 2 with 2-4 mol.% Y203 was tetragonal, while the Z r O 2 with 8 mol.% Y203 was cubic. The M o S i 2 exhibited the tetragonal crystal structure. Similarly, for the (ZrO2-SiC)-MoSi 2 composites, only the Z r O 2 , SiC and MoSi 2 phases were observed. Z r O 2 crystal structures were the same as in the ZrO2-MoSi~ composites. The SiC exhibited an a crystal structure, while the M o S i 2 phase was again tetragonal. These phase analysis results indicate that there was no reaction between the species Z r O 2 , SiC and M o S i 2 at the temperature of 1700 °C employed for composite consolidation.

3.3. Substructures TEM substructures of ZrO2-MoSi 2 and (ZrO2-SiC)M o S i 2 composites containing 0Y203 + Z r O 2 and 2.5Y203 + Z r O 2 compositions are shown in Figs. 2 and 3. A prominent amorphous glassy silica phase was observed in all composites. This glassy phase was most prominent in the areas of Z r O 2 and Z r O 2 - S i C particle clusters. It is also highly likely that MoSi 2 grain boundaries contained a thin grain boundary silica phase. A qualitative TEM chemical analysis of the glassy phase showed it to be essentially SiO2, with a small amount of zirconium, yttrium and molybdenum present. Thus, the glassy phase displayed dissolved elements from some of the composite constituent phases. In Fig. 2, the 20vol.%0Y20 s + Z r O 2 - M o S i 2 c o m p o s i t e exhibited ZrO 2 particles that were monoclinic. Thus, the tetragonal-to-monoclinic phase transformation occurred spontaneously on cooling from the composite fabrication temperature. The M o S i 2 phase in this composite showed a high density of both dislocations and microcracks, which were induced by the spontaneous zirconia transformation [13]. Densities of M o S i 2 phase dislocations and microcracks were lower in the 10vol.%(0Y203 + ZrO2-SiC)LMoSi2 composite.

Fig. 2. TEM substructures of (a) 20vol.%ZrO2-MoSi 2 and (b) (10vol.%ZrO 2-10vol.%SiC)-MoSi 2 composites containing the 0Y~_O3+ ZrOz composition.

This is probably due to the lower volume fraction of zirconia in this composite. In Fig. 3, the 20vol.%2.5Y203 + ZrO2-MoSi 2 composite showed ZrO2 particles which were tetragonal. In this material, the tetragonal-to-monoclinic phase transformation had not occurred spontaneously on cooling. Dislocation densities and microcracking were observed to be low in the MoSi 2 phase [13]. The substructure of the 10vol.%(2.5Y203+ZrO2-SiC)-MoSi 2 composite was similar to that of the ZrO:-MoSi2 composite, except for the presence of the fine SiC particles.

3.4. Room temperature fracture toughness The room temperature indentation fracture toughness of 20vol.%ZrO2-MoSi 2 composites containing various levels of YEO3 stabilizer in the ZrO2 reinforcement is shown in Fig. 4. A similar plot for zirconiatoughened A1203 (ZTA) is also shown [7], for comparison with zirconia-toughened MoSi2 (ZTM).

J.J. Petrovic et al.

262

/

ZrO2-MoSi 2 and (ZrO2~SiC)-MoSi2 composites I

I

I

I

I

I

I

E

I

I

0 2.5 Y ZrO 2

~9 Z "1"

o

F- 4 PURE A I 2 ~ LU CC

p'/

J

V-C..)

_ f_

• • ZrO 2 - (2.5% Y 2 0 3 ) - MoSi 2

•

n" '~ PURE I M°Si2 I 1.1_ 2 20

A Z r O 2 (2.0 MOL% Y203) - AI 2 03 I

I 40

I

I 60

I

I 80

I 100

VOLUME % ZrO 2

Fig. 5. Effects of volume fraction Z r O 2 reinforcement on the room temperature fracture toughness of 2.5Y203 + Z r O 2 particle-MoSi2 matrix composites.

Fig. 3. T E M substructures of (a) 20vol.%ZrO2-MoSi2 and (b) (10vol.%ZrO2-10vol.%SiC)-MoSi 2 composites containing the 2.5Y203 + Z r O 2 composition.

i

i

•

~

i

r

i

i

i

E

I

I

20 Vol % ZrO 2 (P) - MoSi 2 MATRIX x

'-

PURE\

PURE MoSi2 i

2 1

0

I

I

2

I

I

4

I

I

6

I

8

10

MOL% Y2 03 STABILIZER IN ZrO 2 REINFORCEMENT

Fig. 4. Room temperature fracture toughness of 20vol.%ZrO2MoSi 2 composites as a function of amount of Y203 stabilizer in the Z r O 2 reinforcement, and comparison with the behavior of 20voI.%ZrO2-AI203 composites.

As may be seen, the highest toughness for the ZrO2-MoSi 2 composites occurred with unstabilized (0 mol.% Y203)ZrO 2. In this case the composite toughness was three times higher than that of pure MoSi 2, a substantial toughening effect. Partially stabilized (2-4 mol.% Y203) and fully stabilized (8 mol.% Y203) compositions showed lower toughness levels, but still above pure MoSi 2. By way of comparison, the toughness of unstabilized ZTA is only slightly above pure A1203, with maximum toughening occurring in the partially stabilized zirconia range. The toughness of fully stabilized ZTA is reported lower than that of pure A1203. Figure 5 shows effects of volume fraction ZrO2 reinforcement on the room temperature fracture toughness of 2.5Y203 + ZrO2-MoSi2 ZTM composites. Also shown for comparison are volume fraction effects in 2.0Y203 +ZrOE-Al203 ZTA composites. The ZTM composites exhibited a linear relationship between fracture toughness and volume fraction ZrO 2 reinforcement. For the ZTA composites, the behavior roughly parallels that of the Z T M composites up to 45 vol.%, and then deviates toward the somewhat lower value of the 100% 2.0Y203 + ZrO 2. A comparison of indentation fracture toughness as a function of crack length (basically, measured indentation toughness as a function of indentation load at 20, 30 and 50 kgf) for ZrOz-MoSi2 composites and (ZrO2-SiC)-MoSi2 composites is shown in Fig. 6. The slightly lower absolute toughness values for the 2 0 v o I . % 0 Y 2 0 3 + Z r O 2 - M o S i 2 material in comparison with the value in Fig. 4 are attributed to operator judgment in measuring the length of indentation cracks (particularly when significant microcracking is present), since different investigators made these two sets of measurements.

J. J. Petrovic et al.

/

@~PSZ)

7

263

ZrO2-MoSi_, and (ZrO_,-SiC)-MoSi2 composites 700 )

sic-o YPSZ

I)

sic-2.5 YPSZ

(300'

sic-s vPSZ

500 ¸

O-------

0 YPSZ

•

2.5 YPSZ

•............-~- -

8 YPSZ

~5 g~ ,,~ ¢~

10% sic-10% 10 zro2(2.5Ypsz

•

400

[]

.ff

2(2.5YPSZ) [--,

e

4

J

Lr-

30O

20% S~C-10%ZRO2(8¥ ,* 10% SiC-10% ZrO2(8YPSZ

0

200

3

100

2 0.20

i

i

i

I

i

0.25

0.30

0.35

0.40

0.45

0.50 0

300

600

900

1200

1500

Crack Length, mm

Fig. 6. Comparison of the indentation fracture toughness of ZrO2-MoSi2 and (ZrO2-SiC)-MoSi2 composites, as a function of indentation crack length.

The most striking aspect in Fig. 6 is the fact that the fracture toughness of the (0Y203 + ZrOz-SiC)-MoSi2 composite was higher than the toughness of the 0Y203 + ZrOz-MoSi2 composite, despite the fact that the former material contained only 10 vol.% ZrO 2 while the latter contained 20 vol.% Z r O 2. Toughness values for the 2.5Y203+ZrO2-SiC composite were slightly higher than for the 2.55(203 + Z r O 2 composite, while 8Y203+ZrO 2 composite toughnesses were roughly comparable. All composites exhibited a trend of increasing toughness with increasing indentation crack length (increasing indentation load), suggesting R-curve type behavior in these materials. 3.5. B e n d strength vs. temperature

Bend tests were performed at room temperature, 1200 °C, and 1400 °C in air. Figure 7 plots strength vs. temperature for the ZrO2-MoSi 2 and (ZrO2-SiC)M o S i 2 composites. Room temperature strengths are ultimate brittle fracture strengths. Strengths at 1200 °C and 1400°C are yield strengths in bending, since composites exhibited plastic deformation at elevated temperatures. Yield strength in bending was defined as the strength level at 0.2% plastic bending strain. At room temperature, composites containing 2.5Y203 + Z r O z showed the highest strength levels, while composites containing 0Y203 + ZrO2 exhibited the lowest strengths. The 8Y203 +ZrO2 composites were intermediate. At 1200°C, strengths of the

Temperature,C

Fig. 7. Bend strength vs. temperature, for ZrO2-MoSi2 and (ZrO2-SiC)-MoSi2composites. (ZrOE-SiC)-MoSi 2 composites were roughly double those of the ZrO2-MoSi 2 composites, for all ZrO2 compositions. Thus, the presence of the SiC phase produced a significant strengthening effect at 1200 °C. At 1400 °C, the strengths of all composites decreased dramatically to a low level. This decrease is attributed to the presence of the glassy silica phase in the composites.

4. Discussion 4.1. Z r O 2 transformation toughening effects

The present results indicate that ZrO2 produces substantial room temperature transformation toughening effects in MoSi2 based (ZTM) composites. Toughening effects are most pronounced for unstabilized (0 mol.% Y203) ZrO2, in distinct contrast to the behavior of ZTA. In the case of unstabilized ZTA, toughening appears to be associated with microcracking mechanisms [9]. A very intriguing additional aspect occurring in unstabilized ZrO2-MoSi 2 composites is the "pumping" of dislocations into the MoSi 2 matrix as a result of the spontaneous ZrO2 tetragonal-tomonoclinic phase transformation. On cooling from the fabrication temperature (1700 °C), this transformation initiates in the vicinity of 1175 °C [6]. The unstabilized zirconia transformation temperature is above the brittle-to-ductile transition of MoSi2, and so dislocation pumping occurs as a result of the spontaneous

264

J.J. Petrovicet al. / ZrO2-MoSi2and (Zr02-SiC)-MOSi2 composites

Z r O 2 transformation strains. No such phenomena occurs in the ZTA composite system, since A1203

possesses insufficient dislocation plasticity at the unstabilized ZrO2 transformation temperature [9]. It is possible that the increased dislocation density induced in the MoSi2 matrix by the unstabilized ZrO2 may be a contributing factor to the observed toughening effects. For such a mechanism to operate, these induced dislocations would have to be mobile at the MoSi 2 crack tip at room temperature, well below the brittle-toductile transition. Clearly, additional investigations targeted at definitively establishing (or ruling out) such a toughening mechanism are warranted. Present results indicate that crack-tip-induced transformation toughening effects were not yet optimized in current partially stabilized (2-4 mol.% Y203) ZrO 2 ZTM composites. This is in contrast to ZTA composites, where maximum toughening effects occur [7, 9]. It is believed that the lack of sufficient microstructural dispersion of individual partially stabilized Z r O 2 particles in the MoSi 2 matrix of the current composites is responsible for repressing these crack tip transformation effects. A recent investigation has demonstrated that partially stabilized ZTM composites exhibit R-curve behavior, and that such behavior is controlled by the composite critical transformation stress [15]. A high critical stress level observed in current ZTM composites is considered indicative of the inhibition of crack tip transformation toughening effects resulting from microstructural inhomogeneity.

4.2. Effects of combined ZrO2 and SiC reinforcements (ZrO2-SiC)-MoSi z composites were stronger than ZrO2-MoSi 2 composites at 1200°C, and were also tougher materials at room temperature. The strengthening observed at 1200°C is possibly related to the presence of the SiC particulate phase more effectively impeding dislocation motion at this temperature. Significant improvements in elevated temper~iture strength with the presence of an SiC reinforcing phase have certainly been observed in previous work [16]. Present results might suggest that SiC is a more effective elevated temperature strengthening phase than Z r O 2. However, the dispersion homogeneity of the ZrO 2 phase in the ZrO2-MoSi 2 composites was certainly not optimum, and the dispersion of reinforcing particulate phases was clearly significantly improved in the (ZrO2-SiC)-MoSi2 composites. It is possible that once a microstructural condition of isolated, submicron ZrO2 particles in the MoSi 2 matrix is achieved, ZrO 2 particles may prove to be as effective obstacles to dislocation motion as SiC particles. The room temperature fracture toughness levels of unstabilized (0Y203 + Z r O 2 - S i C ) - M o S i 2 and partially stabilized (2.5Y203 +ZrO2-SiC)-MoSi2 composites

were higher than those of their corresponding Z r O 2M o S i 2 composite counterparts, despite the fact that these composites contained half the amount of Z r O 2 phase. It is believed that some of this improvement is related to the improved dispersion of Z r O 2 particles achieved in the (ZrO2-SiC)-MoSi 2 composites in comparison with the Z r O 2 - M o S i 2 composites. However, recent work [17] has shown that a dispersed particulate SiC phase can produce room temperature toughening effects of itself. Thus, in (ZrO2-SiC)-MoSi 2 composites, it appears that Z r O 2 particle and SiC particle toughening effects are certainly at least additive, and may be interactive in some way. One possible interaction mechanism might be related to the thermal expansion mismatch stresses associated with SiC particles enhancing ZrO 2 transformation toughening effects.

4.3. Effects of glassy phase on elevated temperature strength Present results clearly demonstrate the detrimental effects of the presence of a microstructural glassy phase on the elevated temperature mechanical properties of both ZrO2-MoSi2 and (ZrO2-SiC)-MoSi 2 composites. Such detrimental glassy phase effects have also been observed in polycrystalline MoSi 2 [18]. When a glassy phase is present along the grain boundaries of the MoSiz matrix, it acts to induce grain boundary sliding in preference to dislocation motion as the dominant deformation mechanism. In the presence of the glassy phase, grain boundary sliding becomes more dominant with increasing temperature, as the viscosity of the glassy phase decreases. In the present composites, dislocation motion was the probable primary deformation mechanism at 1200 °C, since the grain boundary glassy phase viscosity was high at this temperature. However, at 1400°C the glassy phase viscosity was sufficiently low that grain boundary sliding probably became the dominant deformation mechanism, with associated macroscopic plastic strain occurring at low stress levels. This is the probable reason for the significant reduction in composite strength observed at 1400 °C. 4.4. Aspects for composite mechanical property optimization There are two clear avenues to follow in order to optimize both the high and low temperature mechanical properties of ZrOe-MoSi 2 and (ZrO2-SiC)MoSi 2 composites. The first is elimination of the silica glassy phase and the second is refinement in the microstructural dispersion of reinforcing particulate phases. The glassy silica phase occurs in MoSi 2 composites since it is present on the surfaces of the starting commercial MoSi 2 powders used to fabricate these compo-

J. J. Petrovic et al.

/

ZrO2-MoSi 2 and (ZrO2-SiC)-Mosi 2 composites

sites. Recent work has demonstrated that carbon additions to MoSi 2 are very effective in eliminating the silica glassy phase, via reactions between carbon and silica at elevated fabrication temperatures to form SiC and gaseous CO [18]. Removal of the glassy phase markedly improved the high temperature mechanical properties of polycrystalline MoSi 2. Similar improvements with carbon additions are to be expected for the present composites. The second aspect involves developing fabrication techniques designed to produce a microstructure of isolated, well-dispersed, submicron ZrO2 and SiC particles in the MoSi 2 matrix. This suggests the use and optimization of high energy powder milling techniques such as vibratory milling and mechanical alloying to blend composite constituent powders intimately, and HIP to consolidate these powders at lower temperatures where microstructural coarsening during the consolidation process may be minimized. Improvements in microstructural homogeneity were clearly achieved in present (ZrO2-SiC)-MoSi 2 composites when vibratory milling and HIP consolidation were introduced.

5. Conclusions ZrOz-MoSi2 and (ZrO2-SiC)-MoSi 2 composites were fabricated by hot pressing and hot pressing-HIP at 1700 °C, and their microstructures and mechanical properties evaluated. Under the fabrication conditions, no reactions between ZrO2, SiC, and MoSi 2 were observed. A combination of powder vibratory milling followed by hot pressing-HIP consolidation was observed to improve the microstructural homogeneity of (ZrO2-SiC)-MoSi 2 composites, as compared with ZrOz-MoSi 2 composites which were fabricated by aqueous co-dispersion of powders followed by hot pressing. An amorphous glassy silica phase was observed in all composite materials, as a result of the silica present on the surfaces of the starting commercial MoSi2 powders. The presence of an unstabilized ZrO 2 particulate phase in ZrOz-MoSi 2 (ZTM) composites increased the room temperature indentation fracture toughness by a factor of 3 over that of pure MoSi 2. This is in contrast to the behavior of ZrO2-Al203 (ZTA) composites. For unstabilized ZrO2, microcracking and dislocation pumping were seen in the MoSi 2 matrix. Observed composite toughness levels for partially stabilized and fully stabilized ZrO 2 were lower than those of unstabilized ZrO 2. Microstructural inhomogeneity of the partially stabilized ZrO 2 particles in the MoSi 2 matrix was considered deleterious to the optimum occurrence of crack-tip-induced transformation toughening

265

effects. Fracture toughness of partially stabilized ZrO2-MoSi 2 composites increased linearly with increasing ZrO2 volume fraction. The fracture toughness of unstabilized and partially stabilized (ZrO2-SiC)-MoSi2 composites was greater than that of ZrO2-MoSi 2 composites, and was probably due to a combination of microstructural homogeneity and possible ZrO2-SiC interactive effects. (ZrO2-SiC)-MoSi 2 composites were stronger than ZrO2-MoSi 2 composites at 1200°C. This may be primarily due to the improved microstructural homogeneity of the former composites as compared with the latter, although a possible greater effectiveness of SiC particles over ZrO2 particles to retard elevated temperature dislocation motion in the MoSi 2 matrix cannot be discounted. The strength of all composites was degraded at 1400°C because of the presence of an amorphous glassy silica phase in the MoSi 2 matrix. This glassy phase induced grain boundary sliding to occur at low stress levels at 1400 °C, rather than the operation of dislocation plasticity taking place at higher stress levels. The results of this investigation clearly define two processing routes for the optimization of both the high and low temperature mechanical properties of composites of this type. The first is to eliminate the silica glassy phase, and the second is to refine the microstructural dispersion of the reinforcing particulate phases.

Acknowledgments The authors gratefully acknowledge the Department of Energy-Advanced Industrial Concepts Materials Program for support of the present investigation. Two of the authors (R.K.W. and K.J.M.) additionally acknowledge the Los Alamos National Laboratory Graduate Research Associate Program.

References 1 R. C. Garvie, R. H. Hannink and R. T. Pascoe, Ceramic steel?, Nature (London), 258 (1975) 703. 2 D. L. Porter and A. H. Heuer, Mechanisms of toughening partially stabilized zirconia (PSZ), J. Am. Ceram. Soc., 60 (1977)183. 3 D. L. Porter, A. G. Evans and A. H. Heuer, Transformationtoughening in partially-stabilized zirconia (PSZ), Acta Metall., 27(1979) 1649. 4 A. G. Evans and A. H. Heuer, Review--transformation toughening in ceramics: martensitic transformations in crack-tip stress fields, J. Am. Ceram. Soc., 63 (1980) 241. 5 A. G. Evans and R. M. Cannon, Toughening of brittle solids by martensitic transformations, Acta Metall., 34 (1986) 761. 6 R. Stevens; Zirconia and Zirconia Ceramics, Magnesium Elektron, Twickenham, 2nd edn., 1986. 7 F. F. Lange, Transformation toughening. Part 4: fabrication,

266

8

9 10 11

12 13

J.J. Petrovic et al.

/

ZrO2-MoSi e and (ZrO2-SiC)-MoSi 2 composites

fracture toughness and strength of AI203-ZrO2 composites, J. Mater. Sci., 17(1982)247• K. Tsukuma, K. Ueda and M. Shimada, Strength and fracture toughness of isostatically hot-pressed composites of AI203 and Y203 partially stabilized ZrO 2, J. Am. Ceram. Soc., 68 (1985)C-4. J. Wang and R. Stevens, Review: zirconia-toughened alumina (ZTA) ceramics, J. Mater• Sci., 24 (1989) 3421. N. Claussen, K. L. Weisskopf and M. Ruhle, Tetragonal zirconia polycrystals reinforced with SiC whiskers, J. Am. Ceram. Soc., 69 (1986) 288. M. Backhaus-Ricoult, Oxidation behavior of SiC whisker reinforced alumina-zirconia composites, J. Am. Ceram. Soc., 74(1991) 1793. J. J. Petrovic and R. E. Honnell, Partially stabilized ZrO2 particle-MoSi 2 matrix composites, J. Mater. Sci., 25 (1990) 4453. J. J. Petrovic, R. E. Honnell, T. E. Mitchell, R. K. Wade and

14

15 16 17 18

K. J. McClellan, ZrO 2 reinforced-MoSi 2 matrix composites, Ceram. Eng. Sci. Proc., 12 (1991) 1633. G. R. Anstis, P. Chantikul, B. R. Lawn and D. B. Marshall, A critical evaluation of indentation techniques for measuring fracture toughness: I. Direct crack measurements, J. Am. Ceram. Soc., 64 (1981) 533. A. K. Bhattacharya and J. J. Petrovic, R-curve behavior in ZrO 2 reinforced MoSi 2 composites, J. Am. Ceram. Soc., 75 (1992) 23. J.J. Petrovic and R. E. Honnell, SiC reinforced-MoSi2/WSi2 alloy matrix composites, Ceram. Eng. Sci. Proc., 11 (1990) 734. A. K. Bhattacharya and J. J. Petrovic, Hardness and fracture toughness of SiC particle-reinforced MoSi 2 composites, J. Am. Ceram. Soc., 74(1991)2700. S. Maloy, A. H. Heuer, J. Lewandowski and J. Petrovic, Carbon additions to MoSi2: improved high temperature mechanical properties, J. Am. Ceram. Soc., 74 (1991) 2704.